Day 214 - Friday Fail, raft edition

Here's my favorite fail from this week: a botched print of three Fidget Cubes. It's partially my favorite because I happened to have been sitting right next to the machine when it happened so I could stop it right away before the dreaded "giant blob built up around nozzle" problem kicked in.

It's also my favorite because it helped me to figure something out. I recently switched from the MakerBot acrylic build plate that ships with the Replicator 2 to a glass build plate that I bought on Amazon. When using the acrylic plate, this is the order I would see for platform adhesion (loosest adhesion on the left, most stick-to-it-iveness on the right):

Acrylic-with-tape   <   Acrylic-with-raft   <   Acrylic

Now that I'm using a glass platform, the tape actually makes things adhere better to the platform, not worse, since glass alone is so bad for adhesion. Since I had formerly used rafts to make things stick less to my acrylic platform, it didn't occur to me today to try to use them to make things stick more, even though it is obvious that a raft should make things stick better to tape no matter what the tape is on! With glass platforms, we have this adhesion order (again, loosest on the left):

Glass   <<   Glass-with-tape   <   Glass-with-tape-and-raft

Anyway, long story short, now I'm using Glass-with-tape-and-raft for my prints, and I'm getting a lot fewer "blob" errors. And even better, figuring out this fail helped me to figure out another.

On Day 208 I started printing out a great many 26mm Level 3 Menger cubes. (Side note to @standupmaths - I actually put the package in the mail today, can you believe that? Two months earlier than usual.) Lots of the prints were turning out like Day 207's free-floating half-Menger-attached-to-blob, and no amount of re-leveling the build platform, installing fresh blue tape, or "just trying one more time to see if the problem would magically stop" was fixing the problem. Until we tried printing with a raft - not because the raft made it stick better, but because it turned out that for this model the raft was really really small and didn't cover the base; see the left object in the picture below (shown upside-down so you can see the tiny raft).

Mystery solved! The raft thinks that the first layer of the model is really small. Which means that a tiny dot in the center of the model must be too low: at least for very small sizes like this 26mm cube, a little piece of the bottom center corner of owens' Customizable Menger Sponge sticks down too far and needs to be sliced off.

We went into OpenSCAD and sliced off the very bottom of the model (here "owens_code" is the part of the code where owens' Menger-with-stand object is generated, and we are taking away from that object a cube that sits right under the xy-plane):

difference(){
owens_code();
translate([0,0,-20]) cube(40,center=true);
}

As you can see from the model on the right in the picture above, this change worked; the raft now covers the entire stand and we got a very sturdy print. Thank you once again, failure, for pointing us in the right direction!